Understanding the Puzzle of Primary Health-care Use: Evidence from India Pramod Kumar Sur Asian Growth Research Institute (AGI) and Osaka University [email protected] Abstract In India, households’ use of primary health-care services presents a puzzle. Even though most private health-care providers have no formal medical qualifications, a significant fraction of households use fee-charging private health-care services, which are not covered by insurance. Although the absence of public health-care providers could partially explain the high use of the private sector, this cannot be the only explanation. The private share of health-care use is even higher in markets where qualified doctors offer free care through public clinics; despite this free service, the majority of health-care visits are made to providers with no formal medical qualifications. This paper examines the reasons for the existence of this puzzle in India. Combining contemporary household-level data with archival records, I examine the aggressive family planning program implemented during the emergency rule in the 1970s and explore whether the coercion, disinformation, and carelessness involved in implementing the program could partly explain the puzzle. Exploiting the timing of the emergency rule, state-level variation in the number of sterilizations, and an instrumental variable approach, I show that the states heavily affected by the sterilization policy have a lower level of public health-care usage today. I demonstrate the mechanism for this practice by showing that the states heavily affected by forced sterilizations have a lower level of confidence in government hospitals and doctors and a higher level of confidence in private hospitals and doctors in providing good treatment. JEL Codes: I11, N35, I12, J13 Keywords: Health-care use, family planning, sterilization, confidence in institutions, persistence, India 1 1. Introduction In India, households’ use of primary health-care services presents a puzzle.1 The puzzle is as follows. First, a significant fraction of households uses fee-charging private health-care services, which are not covered by insurance (CPR 2011; International Institute for Population Sciences (IIPS) 2017; Peters et al. 2002).2 Second, major portions of private health-care providers have no formal medical qualifications (Rohde and Viswanathan 1995; Banerjee, Deaton, and Duflo 2004).3 Although public health-care providers are more qualified and offer free services, they have only around 20% of the market share (Muralidharan et al., n.d.).4 Third, while the absence of public health-care facilities or personnel could partly explain the high use of the private sector, this cannot be the only explanation. The private share of health-care use is higher even in markets where qualified doctors offer free care in public hospitals; despite this service, the majority of health-care visits are made to providers with no formal medical qualifications (Das, Holla, et al. 2016). Fourth, within India, there is a considerable variation in the types of health-care usage across states (Peters et al. 2002; Muralidharan et al., n.d.). Why does this paradoxical situation exist in India? This paper examines a plausible reason for the existence of such a paradox. In particular, I question whether the current practice of health-care use in India has historical routes. Combining contemporary household-level data with archival records, I examine the aggressive family planning program implemented in India during the emergency rule in the 1970s and explore 1 Henceforth, I refer to “primary health-care” as the “health-care” for simplicity. 2 For example, India has one of the highest proportions of private health spending anywhere in the world, constituting 82% of all health expenditure. Only five countries (Cambodia, the Democratic Republic of the Congo, Georgia, Myanmar, and Sierra Leone) have a higher dependence on private health financing (Peters et al. 2002). 3 For example, according to the Indian Medical Association, about 1 million unqualified doctors practice allopathic medicine in India. https://www.ima-india.org/ima/left-side-bar.php?pid=291. Accessed on January 28, 2021. 4 The Medical Advice, Quality, and Availability in Rural India (MAQARI) project. http://pubdocs.worldbank.org/en/161151429125257286/pdf/13-Medical-Advice-Quality-and-Availability-in-Rural- India-MAQARI-Karthik-Muralidharan.pdf. Accessed on January 28, 2021. 2 whether the coercion, disinformation, and carelessness involved in implementing the program could partly explain the puzzle. India experienced a brief period of autocratic rule between June 1975 and January 1977.5 This period, popularly known as “the emergency,” which was proclaimed by then Prime Minister Indira Gandhi under the Indian constitution, suspended a wide range of civil liberties. A unique policy that affected the general population during this period was the introduction of an aggressive family planning program through forced sterilization.6 The policy—implemented in April 1976— led to a sharp increase in the number of sterilizations (see Figure 1). About 8.3 million sterilizations were performed in a single year between April 1976 and March 1977, more than three times the previous year’s figure. Historical records, court rulings, and anecdotal evidence suggest that these sterilization targets were accomplished through incentives and disincentives, coercion, disinformation, carelessness, and fear (Shah Commission of Inquiry 1978; Panandiker, Bishnoi, and Sharma 1978). I hypothesize that the forced sterilization policy may have had unintended effects on future health-care usage in India. There are genuine reasons to believe that the policy could have unintended consequences. First, all sterilizations, mostly administered through coercion and disincentives, were performed by government doctors in public hospitals or temporary sterilization camps established by the government. Due to increased pressure, targets to meet, and carelessness, no aftercare was administered, which sometimes led to serious side effects, including death. According to the report published by the Indian Government, 1,778 complaints of deaths related to sterilization were registered. Second, false information was delivered by public health-care 5 The autocratic rule (the emergency) officially ended in March 1977. However, it was substantially relaxed in January 1977. 6 Henceforth, I refer to “the aggressive family planning program through forced sterilization” as the “forced sterilization policy” or “the policy” for simplicity. 3 workers to motivate individuals to be sterilized. In a survey of four Indian states (Bihar, Madhya Pradesh, Punjab, and Uttar Pradesh) during the forced sterilization period, Panandiker, Bishnoi, and Sharma (1978) summarize the types of disinformation provided to motivate sterilization acceptors: What was often told was that sterilization, vasectomy or tubectomy, is a simple, quick and safe operation which stops child birth permanently… To the more circumspect of the prospective clients, it was also quietly added that in case of need for a child-birth later it could be reversed also. Nobody explained how an operation is performed, in what manner it stops the conception and what its consequences are to the health of a person … As the program was generally time and target bound, their mission was “Quick Catch” rather than to carry conviction. (p. 104) Considering these insights, I examine the consequences of the forced sterilization policy on India’s future health-care practice. To test my hypothesis, I use data from India’s national representative National Family and Health Survey in 2015–16 (NFHS-4) to examine the sources of households’ health-care use. To measure the exposure to the forced sterilization policy, I digitize and use various state-level sterilization performance statistics from the historical yearbooks published by the Ministry of Health and Family Planning, Government of India. I find that higher exposure to the forced sterilization policy is associated with lower use of public health-care facilities today. My results are robust to a variety of controls and a number of alternative measures of exposure to the forced sterilization policy. After establishing that the forced sterilization policy has a negative association with the use of public health-care facilities today, I turn to the task of addressing concerns of reverse 4 causality and omitted variable bias using an instrumental variable (IV) approach. To identify the causal impact, I need an instrument that exogenously determines the sterilization performance during this period. For this, I exploit the unique history of the implementation of the forced sterilization policy and use distance from New Delhi to state capitals as an instrument. This instrument is developed by and empirically tested in Sur (2021) to examine the impact of the forced sterilization policy on lower vaccination rate in India. The instrument is constructed considering the well know insight from the emergency period that the forced sterilization policy was aggressively undertaken owing to the active role played by Sanjay Gandhi, the son of the Prime Minister Indira Gandhi (Gwatkin 1979; Indian National Congress 2011; Nayar 2013; Chandra 2017). Due to Mr. Gandhi’s personal influence, forced sterilization was aggressively undertaken in the northern parts of India, and distance from New Delhi, which was previously irrelevant, emerged as an important determinant of excess sterilizations